Exploiting Programmable Network Interfaces for Parallel Query Execution in Workstation Clusters

Workstation clusters equipped with high performance interconnect having programmable network processors facilitate interesting opportunities to enhance the performance of parallel application run on them. In this paper, we propose schemes where certain application level processing in parallel database query execution is performed on the network processor. We evaluate the performance of TPC-H queries executing on a high end cluster where all tuple processing is done on the host processor, using a timed Petri net model, and find that tuple processing costs on the host processor dominate the execution time. These results are validated using a small cluster. We therefore propose 4 schemes where certain tuple processing activity is offloaded to the network processor. The first 2 schemes offload the tuple splitting activity - computation to identify the node on which to process the tuples, resulting in an execution time speedup of 1.09 relative to the base scheme, but with I/O bus becoming the bottleneck resource. In the 3rd scheme in addition to offloading tuple processing activity, the disk and network interface are combined to avoid the I/O bus bottleneck, which results in speedups up to 1.16, but with high host processor utilization. Our 4th scheme where the network processor also performs apart of join operation along with the host processor, gives a speedup of 1.47 along with balanced system resource utilizations. Further we observe that the proposed schemes perform equally well even in a scaled architecture i.e., when the number of processors is increased from 2 to 64

Mobile node authentication using key distribution scheme in wireless sensor networks

We consider the problem of secure communication in mobile Wireless Sensor Networks (WSNs). Achieving security in WSNs requires robust encryption and authentication standards among the sensor nodes. Severe resources constraints in typical Wireless Sensor nodes hinder them in achieving key agreements. It is proved from past studies that many notable key management schemes do not work well in sensor networks due to their limited capacities. The idea of key predistribution is not feasible considering the fact that the network could scale to millions. We prove a novel algorithm that provides robust and secure communication channel in WSNs. Our Double Encryption with Validation Time (DEV) using Key Management Protocol algorithm works on the basis of timed sessions within which a secure secret key remains valid. A mobile node is used to bootstrap and exchange secure keys among communicating pairs of nodes. Analysis and simulation results show that the performance of the DEV using Key Management Protocol Algorithm is better than the SEV scheme and other related work.

A fractional order proportional integral controller for path following and trajectory tracking of miniature air vehicles

In this paper, a fractional order proportional-integral controller is developed for a miniature air vehicle for rectilinear path following and trajectory tracking. The controller is implemented by constructing a vector field surrounding the path to be followed, which is then used to generate course commands for the miniature air vehicle. The fractional order proportional-integral controller is simulated using the fundamentals of fractional calculus, and the results for this controller are compared with those obtained for a proportional controller and a proportional integral controller. In order to analyze the performance of the controllers, four performance metrics, namely (maximum) overshoot, control effort, settling time and integral of the timed absolute error cost, have been selected. A comparison of the nominal as well as the robust performances of these controllers indicates that the fractional order proportional-integral controller exhibits the best performance in terms of ITAE while showing comparable performances in all other aspects.

Primary cosmic-ray positrons and negatrons in 1968 at energies between 11 and 204 MeV

The cosmic-ray positron and negatron spectra between 11 and 204 MeV have been measured in a series of 3 high-altitude balloon flights launched from Fort Churchill, Manitoba, on July 16, July 21, and July 29, 1968. The detector system consisted of a magnetic spectrometer utilizing a 1000-gauss permanent magnet, scintillation counters, and a lucite Čerenkov counter.

Launches were timed so that the ascent through the 100 g/cm² level of residual atmosphere occurred after the evening geomagnetic cutoff transition. Data gathered during ascent are used to correct for the contribution of atmospheric secondary electrons to the flux measured at float altitude. All flights floated near 2.4 g/cm².

A pronounced morning intensity increase was observed in each flight. We present daytime positron and negatron data which support the interpretation of the diurnal flux variation as a change in the local geomagnetic cutoff. A large diurnal variation was observed in the count rate of positrons and negatrons with magnetic rigidities less than 11 MV and is evidence that the nighttime cutoff was well below this value.

Using nighttime data we derive extraterrestrial positron and negatron spectra. The positron-to-total-electron ratio which we measure indicates that the interstellar secondary, or collision, source contributes ≾50 percent of the electron flux within this energy interval. By comparing our measured positron spectrum with the positron spectrum calculated for the collision source we derive the absolute solar modulation for positrons in 1968. Assuming negligible energy loss during modulation, we derive the total interstellar electron spectrum as well as the spectrum of directly accelerated, or primary, electrons. We examine the effect of adiabatic deceleration and find that many of the conclusions regarding the interstellar electron spectrum are not significantly altered for an assumed energy loss of up to 50 percent of the original energy.

Étude hydrologique du plateau continental Ivoirien

The authors give a picture of the average seasonal hydrographic situations over the Ivorian continental shelf using data provided by 26 cruises carried out from July 1969 to January 1972. They study meteorological conditions and the mechanism of setting of different types of hydrographic seasons defined as follows: a cold period related to an upwelling created by winds July to earlier October and a warm period divided in 2 parts in relation with haline variations: a low salinity period in November and December, and a high salinity period from January to May; this one sometimes cut off by short-timed drops in the temperature. Then precisions are given about seasonal and geographical variations using space-time diagrams: last, depth and intensity of the thermocline are examined.

Movements, behavior, and habitat selection of bigeye tuna (Thunnus obesus) in the eastern equatorial Pacific, ascertained through archival tags

Ninety-six bigeye tuna (88– 134 cm fork length) were caught and released with implanted archival (electronic data storage) tags near fish-aggregating devices (FADs) in the equatorial eastern Pacific Ocean (EPO) during April 2000. Twenty-nine fish were recaptured, and the data from twenty-seven tags were successfully downloaded and processed. Time at liberty ranged from 8 to 446 days, and data for 23 fish at liberty for 30 days or more are presented. The accuracy in geolocation estimates, derived from the light level data, is about 2 degrees in latitude and 0.5 degrees in longitude in this region. The movement paths derived from the filtered geolocation estimates indicated that none of the fish traveled west of 110°W during the period between release and recapture. The null hypothesis that the movement path is random was rejected in 17 of the 22 statistical tests of the observed movement paths. The estimated mean velocity was 117 km/d. The fish exhibited occasional deep-diving behavior, and some dives exceeded 1000 m where temperatures were less than 3°C. Evaluations of timed depth records, resulted in the discrimination of three distinct behaviors: 54.3% of all days were classified as unassociated (with a floating object) type-1 behavior, 27.7% as unassociated type-2 behavior, and 18.7% as behavior associated with a floating object. The mean residence time at floating objects was 3.1 d. Data sets separated into day and night were used to evaluate diel differences in behavior and habitat selection. When the fish were exhibiting unassociated type-1 behavior (diel vertical migrations), they were mostly at depths of less than 50 m (within the mixed layer) throughout the night, and during the day between 200 and 300 m and 13° and 14°C. They shifted their average depths in conjunction with dawn and dusk events, presumably tracking the deep-scattering layer as a foraging strategy. There were also observed changes in the average nighttime depth distributions of the fish in relation to moon phase.

Optical characteristics of InAs quantum dots on GaAs matrix by using various InGaAs structures

The effects of various InGaAs layers on the structural and optical properties of InAs self-assembled quantum dots (QDs) grown by molecular-beam epitaxy ( MBE) were investigated. The emission wavelength of 1317 nm was obtained by embedding InAs QDs in InGAs/GgAs quantum well. The temperature-dependent and timed-resolved photoluminescence (TDPL and TRPL) were used to study the dynamic characteristics of carriers. InGaAs cap layer may improve the quality of quantum dots for the strain relaxation around QDs, which results in a stronger PL intensity and an increase of PL peak lifetime up to 170 K. We found that InGaAs buffer layer may reduce the PL peak lifetime of InAs QDs, which is due to the buffer layer accelerating the carrier migration. The results also show that InGaAs cap layer can increase the temperature point when, the thermal reemission and nonradiative recombination contribute significantly to the carrier dynamics.

基于Yices对时间自动机的有界模型检测

随着工业化程度的提高，对模型的检测方法受到越来越大的重视。通常的检测方法有推理验证和模型检测。而相对于推理验证，模型检测由于其高度自动化的检测过程，在工业界有着更广泛的应用。 模型检测自从概念雏形开始之时，就受到状态爆炸问题的困扰。为了解决这个问题，不同的技术被提了出来。McMillan提出利用OBDD的符号模型检测方法在一定程度上解决了这个问题。另外，本文所主要讲述的有界模型检测，经大量的实践证明，对符号模型检测也是一个很好的补充。 由于问题的特殊性，科学家提出来不同的模型来描述不同情况的系统。由于时间自动机能够很好地描述异步系统，而且这种系统广泛地存在于现实生活中，对时间自动机这种特殊模型的验证方法的研究变得很有必要。另外，由于时间自动机带有实数域的时钟变量，这导致了时间自动机有一个无限域的状态迁移图。为了利用模型检测的方法对其进行验证，需要对时间变量进行预处理。一般的方法是把时钟所对应的时钟区或时钟域根据等价性，化为有限的域，相应地，把时间自动机转化为有限的状态迁移图。 为了避免在对时间自动机有界模型检测过程中对变量进行布尔编码以及对时间自动机模型中的时钟进行预处理，本文给出一个利用SMT工具进行的对时间自动机进行有界模型检测的方法。该方法的主要优点是无需将时间自动机中的时钟进行预处理，也不需要将模型中的变量进行布尔编码，只需将时间自动机转变为SMT工具可解的逻辑公式，利用SMT的高效求解来进行模型检测。实验结果表明，对于某些可达性性质的验证，这种方法的效率有一定的优势。

时间符号迁移图及其互模拟判定

引入时间符号迁移图的概念，作为既涉及通讯又具有实时性的并发系统的模型。该文给出了这种迁移图时间互模拟的算法，并证明了该算法的正确性。

一种基于时间自动机的实时系统测试方法

基于时间自动机(timedautomata,简称TA)的一种变体——时间安全输入/输出自动机(timedsafetyinput/outputautomata,简称TSIOA),提出了一种实时系统测试方法.该方法首先将时间安全输入/输出自动机描述的系统模型转换为不含抽象时间延迟迁移的稳定符号状态迁移图(untimedstabletransitiongraphofsymbolicstate,简称USTGSS);然后采用基于标号迁移系统(labeledtransitionsystem,简称LTS)的测试方法来静态生成满足各种结构覆盖标准的包含时间延迟变量迁移动作序列;最后,给出了一个根据迁移动作序列构造和执行测试用例的过程,该过程引入了时间延迟变量目标函数,并采用线性约束求解方法动态求解迁移动作序列中的时间延迟变量.

基于时间自动机的测试用例自动生成及执行方法

为了实现对时间自动机模型的测试,采用符号状态拆分算法对时间自动机的状态空间进行等价划分,以得到最简稳定符号状态转移图,并将其中的抽象时间延迟转移替换为时间延迟变量;针对系统中每个时间自动机建立各自的符号状态转移图,再采用基于符号迁移系统的测试方法分别生成相应的转移动作序列;最后通过对这些序列进行组合产生系统的测试用例,为了执行测试用例,利用TTCN-3的多PTC并发执行能力来实施测试。

面向媒体时序描述的带时间自动机的自动构造方法

依据有穷状态自动机模型，面向程序规范的并发系统和分布式系统测试方法的研究已经取得许多结果．由于特殊的实时和同步要求，这些结果不能直接应用于分布式多媒体软件系统的测试．为此，作者提出一种面向媒体对象时序描述的带时间自动机（Ｔｉｍ ｅｄ ａｕｔｏｍ ａｔａ）的自动构造方法，根据带时间自动机，对分布式多媒体软件系统进行非确定性测试时，可以较容易地判断运行结果正确与否；在进行确定性测试时，可以辅助自动生成测试用例

对时间自动机进行空性检测的过程和改进

为了提高对时间自动机进行空性检测的效率,研究了使用基于时钟区域(zone)的符号化方法和抽象对时间自动机进行空性检测,提出了针对时间自动机自身特点对检测过程进行改进的方法。通过使用基于zone的符号化表示方法和抽象,一个符号化状态表示显式的状态的集合,时间自动机的状态空间会显著缩小,不同的抽象方法对状态空间有不同的效果。符号化状态间不仅有相等关系还有包含关系,通过判断这种包含关系可以尽早的找到接收路径和避免不必要的状态展开从而提高空性检测的效率。实现了改进的检测过程,对一些例子进行了数据比较,取得了较好的实验结果。